3DCGRU: THREE-DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK-BASED BIDIRECTIONAL GATED RECURRENT UNIT MODEL FOR LUNG CANCER CLASSIFICATION

Lung cancer and colorectal cancer are leading causes of U.S. cancer mortality. Because mortality rates for many cancers vary by socioeconomic characteristics, we used area socioeconomic indices to examine patterns in U.S. lung and colorectal cancer mortality between 1950 and 1998. A factor-based area socioeconomic index was linked to 1950-1998 county mortality data to generate annual lung and colorectal cancer mortality rates for each area socioeconomic group. Joinpoint regression analysis was used to model and identify statistically significant changes in the mortality trends. Area socioeconomic patterns in U.S. lung cancer mortality changed dramatically between 1950 and 1998. Men aged 25-64 years and those aged 65 years or older in higher socioeconomic areas generally had higher lung cancer mortality than did those in lower socioeconomic areas during 1950-1964 and 1950-1980, respectively. Area socioeconomic differences in lung cancer mortality began to reverse and widen by the early 1970s for younger men and by the mid-1980s for older men. In 1998, lung cancer mortality was 56% (95% confidence interval [CI] = 49% to 64%) higher for younger men and 38% higher (95% CI = 34% to 43%) for older men in the lowest area socioeconomic group than for the same age groups in the highest area socioeconomic group. Lung cancer mortality among older women in all socioeconomic groups increased sevenfold to eightfold between 1950 and 1998, with higher mortality in higher area socioeconomic groups. The positive socioeconomic gradient in colorectal cancer mortality diminished substantially over time. Although colorectal cancer mortality among women in all area socioeconomic groups showed a consistent downward trend, colorectal cancer mortality among men in low area socioeconomic groups, but not in high area socioeconomic groups, showed an upward trend. Socioeconomic gradients in male lung cancer mortality reversed between 1950 and 1998, and those in colorectal cancer mortality narrowed over that time. Area measures may be useful for monitoring socioeconomic disparities in cancer mortality and for identifying areas for potential cancer control interventions.

Background:The participation rate is one of the most important indexes in the cancer screening. Historically in Japan, each local government has developed their own equations to calculate the subjects for population-based screening, which were different from each other, and therefore the participation rates of screening were not comparable. Recently, local governments were ordered to use the standardized equation in reporting data, which made it possible to compare the participation rates of cancer screening nationwide for the first time. We therefore investigated the correlation between the prefectural lung cancer mortality and several indicators of lung cancer screening.Methods:The prefectural participation rates of lung, gastric and colonic cancer screening, test positive rates, attendance rates for further examination, lung cancer detection rates and positive predictive values of lung cancer screening were collected from “Cancer Registration and Statistics” of the National Cancer Research Center website. The age-adjusted lung, gastric and colonic cancer mortality rates, smoking rates were also collected. The EZR software program was used for statistical analyses.Results:The participation rates of lung cancer screening had a strong positive correlation with the participation rates of gastric/colonic cancer screening (P<0.001). The prefectural lung cancer mortality rates had a moderate to weak negative correlation with the participation rates of lung cancer screening (P=0.009). A little correlation was noted between other quality assurance indicators of lung cancer screening and lung cancer mortality rates.Conclusion:These results suggested that participating in lung cancer screening might help reduce lung cancer mortality rates in some extent.

Among women, lung cancer mortality rates have surpassed those for breast cancer in several countries. This reflects the breast cancer mortality declines due to access to screening and effective treatment alongside the entrance of certain countries in stages of the tobacco epidemic in which smoking becomes more prevalent in women. In this study, we project lung and breast cancer mortality until 2030 in 52 countries. Cancer mortality data were obtained from the WHO Mortality Database. Age-standardized mortality rates (ASMR), per 100,000, were calculated (direct method) for 2008 to 2014 and projected for the years 2015, 2020, 2025, and 2030 using a Bayesian log-linear Poisson model. In 52 countries studied around the world, between 2015 and 2030, the median ASMR are projected to increase for lung cancer, from 11.2 to 16.0, whereas declines are expected for breast cancer, from 16.1 to 14.7. In the same period, the ASMR will decrease in 36 countries for breast cancer and in 15 countries for lung cancer. In half of the countries analyzed, and in nearly three quarters of those classified as high-income countries, the ASMR for lung cancer has already surpassed or will surpass the breast cancer ASMR before 2030. The mortality for lung and breast cancer is higher in high-income countries than in middle-income countries; lung cancer mortality is lower in the latter because the tobacco epidemic is not yet widespread. Due to these observed characteristics of lung cancer, primary prevention should still be a key factor to decrease lung cancer mortality.Significance: The mortality for lung and breast cancer is projected to be higher in high-income countries than in middle-income countries, where lung cancer mortality is expected to surpass breast cancer mortality before 2030. Cancer Res; 78(15); 4436-42. ©2018 AACR.

Informing Patient Surveillance for the Growing Number of Survivors of Lung Cancer

The Associations of Aspirin, Statins, and Metformin With Lung Cancer Risk and Related Mortality: A Time-Dependent Analysis of Population-Based Nationally Representative Data

BackgroundIn the US, lung cancer burden is greater in counties that are either rural or in persistent poverty. This study examined lung cancer risk (e.g., smoking), incidence, and mortality across four county types defined by cross-classification of rurality and persistent poverty.MethodsWe conducted a secondary analysis of county characteristics and lung cancer risk, incidence and mortality. We used data from USDA to classify counties according to rurality (using rural–urban continuum codes) and persistent poverty (i.e., 20% + of residents living below the poverty line for 30 + years). We used publicly-available data to calculate mean county-level prevalence of smoking among adults (in 2019), lung cancer incidence (2015–2019), and lung cancer mortality (2015–2019) across county types. Beta and binomial regression models assessed differences in smoking, lung cancer incidence, and lung cancer mortality by rurality and persistent poverty.ResultsAmong U.S. counties, 1,115 were urban, non-persistent poverty, 1,675 were rural, non-persistent poverty, 52 were urban, persistent poverty, and 301 were rural, persistent poverty. Smoking, lung cancer incidence, and lung cancer mortality were higher in rural counties and in persistent poverty counties than in their comparison counties. Counties that were both rural and persistent poverty had the highest rates of smoking, lung cancer incidence, and lung cancer mortality. Persistent poverty and rurality interacted in their relationship with smoking prevalence (p < 0.01), and lung cancer mortality (p < 0.10).ConclusionsSmoking, lung cancer incidence, and lung cancer mortality are highest in counties that are both rural and persistent poverty, suggesting an urgent need to develop targeted lung cancer interventions in these communities.

WITHIN THE SPACE OF SEVERAL MONTHS, 2 VERY large randomized trials of screening for lung cancer have reported their findings, which fortunately complement one another. The National Lung Screening Study (NLST) found that annual lowdose computed tomography (CT) reduced lung cancer mortality by 20% relative to annual chest radiography. In this issue of JAMA, investigators from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Randomized Trial report that annual screening chest radiography does not reduce lung cancer mortality relative to no screening. Should clinicians infer that screening with low-dose CT reduces lung cancer mortality by 20% relative to no screening? This editorial addresses that question and several other aspects of the PLCO trial. Why would the National Cancer Institute sponsor a large trial of screening for lung cancer with chest radiography? Although 6 randomized trials, most of them published in the 1980s, found no evidence that screening radiography reduced lung cancer mortality, the control group received screening chest radiography in all but the Mayo Lung Project. This study had important protocol deviations and was relatively small (9211 participants randomized; 366 cancers detected). So the body of evidence was inconclusive according to the investigators who designed the PLCO trial. The PLCO trial measured the effect of a package of screening interventions aimed at preventing death from 4 cancers. Patients were individually randomized to a usual care group or to an intervention group that was screened periodically for prostate, lung, colorectal, and ovarian cancer for 3 years and then monitored for PLCO cancers (a stop-screen design). Recruiting targeted the US general population aged 55 through 74 years. The study was designed to have a 90% probability of detecting a 10% reduction in lung cancer mortality. Over 8 years, 77 445 participants were randomized to screening and 77 456 to usual care. Half of the participants were ever-smokers; 10% were current smokers. Lung cancer mortality, the primary end point, was 14.0 per 10 000 person-years of follow-up in the intervention group and 14.2 in the control group (rate ratio, 0.99; 95% CI, 0.87 to 1.22). In high-risk patients who met the NLST eligibility criteria, the outcome was similar except for higher lung cancer mortality rates. The PLCO trial shows that a short-term chest radiography screening program has no effect on lung cancer mortality. The only potential concern about the validity of this conclusion is the reporting of follow-up contact with the trial participants. The authors’ diagram of the flow of participants through the trial (Figure 1) does not state the number of participants in the usual care group that the authors were unable to contact during follow-up. Differential ascertainment of lung cancer mortality and, especially, incidence could occur if follow-up rates were unequal in the screening and usual care groups for reasons linked to lung cancer incidence and mortality. Most of the 1696 cancers were interval cancers (n=198), arose in patients who were never screened (n=193), or arose in patients who had completed 3 rounds of screening (n=998). These far outnumbered the screen-detected cancers (n=307). The large number of deaths from cancers diagnosed after screening is a reminder of the transitory benefit from a short-term program of screening for lung cancer. The best test of lung cancer screening in high-risk individuals would be a trial of lifelong screening. With a stop-screen design, the true effect of screening is unclear because of uncertainty about how long to monitor patients after screening. At one extreme, stopping right after the last screen will miss cancers that screening did not detect but that would have been diagnosed if monitoring had been extended into the postscreening period. Stopping monitoring too soon, therefore, may overestimate the effect of screening by covering up some of its failures. At the other extreme, extending the period of monitoring long past the time when the last cancer missed by screening would have been diagnosed will underestimate the

BackgroundPrior research has not examined the connection between the quality of macronutrients and the occurrence as well as fatality rates of lung cancer (LC). Consequently, to delve deeper into the correlations between macronutrient quality and the likelihood of developing LC, we carried out an extensive, long-term prospective cohort study of 101,755 American adults from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.MethodsOur research cohort comprised 154,887 adults, aged between 55 and 74, who were enrolled from 10 screening facilities across the United States. The macronutrient quality index (MQI) was derived from participants’ responses to a dietary history questionnaire (DHQ). To quantify the strength and precision of the relationships between MQI and the incidence as well as mortality of LC, we employed Cox proportional hazards regression modeling to estimate hazard ratios (HRs) alongside their corresponding 95% confidence intervals (CIs). Additionally, we conducted subgroup analyses to scrutinize whether the observed link between MQI and LC risk was subject to modification by potential confounding variables. To reinforce the reliability of our results, sensitivity analyses were also carried out.ResultsOver an average follow-up period spanning 8.82 ± 1.95 years (accumulating to 897,809 person-years of observation), we recorded 1,706 LC diagnoses, encompassing 1,464 cases of non-small cell lung cancer (NSCLC) and 242 cases of small cell lung cancer (SCLC). Additionally, there were 1,217 deaths attributed to LC, with 1,005 NSCLC-related and 212 SCLC-related fatalities. Our results demonstrate a distinct, statistically significant inverse association between a higher MQI and both a reduced incidence (HR Q4 vs. Q1: 0.65; 95% CI: 0.56–0.76; p < 0.001 for trend) and decreased mortality (HR Q4 vs. Q1: 0.71; 95% CI: 0.60–0.84; p < 0.001 for trend) of LC. This inverse relationship held true for both NSCLC and SCLC subtypes. The robustness of the associations between MQI and the incidence as well as mortality of LC was solidly affirmed through sensitivity analyses.ConclusionOur research outcomes imply that prioritizing the intake of higher-quality macronutrients could serve as a viable strategy to mitigate LC risk within the American population.

Forecasting PM2·5 induced lung cancer mortality and morbidity at county-level in China using satellite-derived PM2·5 data from 1998 to 2016: a modelling study

Background: Prospective evidence on the association between diet patterns and lung cancer risk is limited, particularly in older women. We investigated whether commonly used diet quality indices - Healthy Eating Index 2010 (HEI), Alternate Healthy Eating Index 2010 (AHEI), alternate Mediterranean Diet (aMED) and Dietary Approaches to Stop Hypertension (DASH) - were associated with lung cancer incidence and mortality in the Women's Health Initiative Observational Study (WHI OS). Methods: Self-reported diet intake (food frequency questionnaires) and relevant covariate information was obtained from 86,090 race-ethnically diverse postmenopausal women aged 50-79 years at baseline (1993-1998). During a median follow-up of approximately 17 years, 1,491 lung cancer cases and 1,393 lung cancer deaths were documented. We used cox proportional hazards regression models to estimate hazards risks and 95% confidence intervals [HR (95% CI)] after controlling for age, race/ethnicity, education, body mass index, smoking, physical activity and energy intake. Results and conclusions: Diet quality indices were not associated with lung cancer incidence overall but all four dietary indices showed a protective association (highest versus lowest quintile) against squamous cell carcinoma cases (12.8% of total lung cancer cases) [HEI: 0.56 (0.33-0.96), AHEI: 0.42 (0.24-0.76), aMED: 0.65 (0.39-1.08), DASH: 0.56 (0.32-0.97)]. Diet quality indices were not associated with lung cancer mortality overall or when stratified by histological type, except for HEI with lung cancer mortality overall [fourth versus lowest quintile: 0.82 (0.69-0.99)]. Smoking status did not modify the association between diet quality indices and lung cancer incidence or mortality. Further studies in other populations may help elucidate the relationship between diet patterns and lung cancer, especially by histological type. Citation Format: Ajay A. Myneni, Gary A. Giovino, Amy E. Millen, Michael J. LaMonte, Jean Wactawski-Wende, Marian L. Neuhouser, Jiwei Zhao, James M. Shikany, Lina Mu. Indices of diet quality and risk of lung cancer incidence and mortality in the Women's Health Initiative Observational Study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5261.

The incidence and mortality rates of lung cancer have been increasing in many countries. However, the true effectiveness of screening for lung cancer is still controversial. This study aimed to examine the growth pattern of lung cancer and to evaluate the efficacy of screening. The authors linked the records of annual radiologic screening to cancer registry data and conducted a retrospective follow-up study of radiographs in all patients with lung cancer arising in a population undergoing screening. Among a total of 305,934 participants, screening detected 206 lung cancers, 103 of which were Stage I disease. Seventy-one of the 131 adenocarcinomas were Stage I, and 58% of them showed evidence of cancer for 2 years on a retrospective review of radiographs. Presentation as small faint lesions overlapping the normal chest structures delayed the early detection of adenocarcinoma. The overall sensitivity of screening was 70%, 52% for squamous cell carcinoma and 50% for small cell carcinoma. Rapidly growing Stage II-IV tumors without retrospective evidence of cancer on previous radiographs accounted for most of the cancers detected during the intervals between screening. Both the low detectability of Stage I adenocarcinoma and the late recognition of rapidly growing small cell and squamous cell carcinomas reduced the effectiveness of screening. More effective imaging methods and an antismoking campaign are required to reduce lung cancer mortality.

Objective: Lung cancer (LC) is one of the most common cancers and the main causes of cancer mortality in the world. The aim of this study was to investigate the epidemiology of lung cancer incidence and mortality in 185 countries and its relationship with the HDI index in 2018. Materials and Methods: This is a descriptive-analytic study conducted on the extraction of incidence data and mortality rates of cancers from Cancer World Bank in 2018. The incidence and mortality rates and lung cancer distribution maps were drawn for World countries. To analyze the data, the correlation and regression tests were used to evaluate the correlation between the incidence and mortality with HDI. The statistical analysis was conducted by Stata-14 and the significance level was estimated at the level of 0.05. Results: With 209,386 new cases (12.22 per 100,000) and 1,761,007 deaths (19.88 per 100,000), lung cancer has the highest incidence and mortality rate in the world. The highest incidence rate (56.7 per 100,000) and mortality (44.4 per 100,000) of lung cancer were in Hungary. According to a projection, the incidence and mortality rate of lung cancer are expected to increase from 2018 to 2040, and the results showed that there was a positive and significant correlation between incidence (R=0.724, p 0.05). Conclusions: With the highest global incidence, motility, and an upward trend by 2040, lung cancer has a considerable global importance, and the human development index (HDI) can be an important factor in reducing the incidence and mortality of patients.

The serious ambient fine particulate matter (PM2.5) is one of the key risk factors for lung cancer. However, existing studies on the health effects of PM2.5 in China were less considered the regional transport of PM2.5 concentration. In this study, we aim to explore the association between lung cancer and PM2.5 and then forecast the PM2.5-induced lung cancer morbidity and mortality in China. Ridge regression (RR), partial least squares regression (PLSR), model tree-based (MT) regression, regression tree (RT) approach, and the combined forecasting model (CFM) were alternative forecasting models. The result of the Pearson correlation analysis showed that both local and regional scale PM2.5 concentration had a significant association with lung cancer mortality and morbidity and compared with the local lag and regional lag exposure to ambient PM2.5; the regional lag effect (0.172~0.235 for mortality; 0.146~0.249 for morbidity) was not stronger than the local lag PM2.5 exposure (0.249~0.294 for mortality; 0.215~0.301 for morbidity). The overall forecasting lung cancer morbidity and mortality were 47.63, 47.86, 39.38, and 39.76 per 100,000 population. The spatial distributions of lung cancer morbidity and mortality share a similar spatial pattern in 2015 and 2016, with high lung cancer morbidity and mortality areas mainly located in the central to east coast areas in China. The stakeholders would like to implement a cross-regional PM2.5 control strategy for the areas characterized as a high risk of lung cancer.

Some studies have suggested a protective effect of antioxidant nutrients on lung cancer. Observational epidemiological studies suggest an association between higher dietary levels of fruits and vegetables containing beta carotene and a lower risk of lung cancer. To determine whether vitamins, minerals and other potential agents, alone or in combination, reduce incidence and mortality from lung cancer in healthy people. The electronic databases MEDLINE (1966-july 2001), EMBASE (1974-july 2001) and the Cochrane Controlled Trial Register (CENTRAL, Issue 3/2001) and bibliographies were searched. In addition authors of included studies were contacted to identify potentially eligible published and unpublished trials. Included studies were randomised controlled clinical trials comparing different supplements or comparing supplements with placebo, administered to healthy people with the aim of preventing lung cancer. Three reviewers independently selected the trials to be included in the review and assessed the methodological quality of each trial, and two extracted data using a standardised form. For each study, relative risk and 95% confidence limits were calculated for dichotomous outcomes. Four studies were eligible for inclusion. All were population based trials, including a total of 109,394 participants. Two studies included smokers, one included workers exposed to asbestos and two studies were carried out in health professionals. A group of participants with no known risk factors for lung cancer was included in the study sample of two trials. Beta-carotene was evaluated in all trials, alone or combination with alpha-tocopherol or retinol, and one study tested alpha-tocopherol alone. Duration of treatment varied from 2 to 12 years and follow-up was from two to five years. All trials had a placebo group. For people with risk factors for lung cancer no reduction in lung cancer incidence or mortality was found in those taking vitamins alone compared with placebo (incidence of lung cancer: RR 0.98, 95% CI 0.81-1.19; lung cancer mortality: RR 0.93, 95% CI 0.73-1.19). For people with no known risk factors of lung cancer, none of the vitamins or their combinations appeared to have any effect. Combined data from three studies showed a non-statistically significant increased risk of lung cancer incidence (RR 1.11, 95% CI 0.94-1.33) and mortality (RR 1.05, 95% CI 0.87-1.28) for beta-carotene alone at pharmacological doses in groups with risk factors for lung cancer. When beta-carotene was combined with retinol, data from a single study showed that there was a statistically significant, increased risk of lung cancer incidence (RR 1.42, 95% CI 1.13-1.80) and mortality (RR 1.75, 95% CI 1.29-2.38) in people with risk factors for lung cancer who took both vitamins compared with those who took placebo. Data from also from one study showed that the combination of beta-carotene with alpha-tocopherol in people with risk factors for lung cancer was associated with a non-statistically significant increased risk of lung cancer incidence (RR 1.16, 95% CI 0.96-1.39) and mortality (RR 1.15, 95% CI 0.91-1.45). No effect was observed for total cancer incidence, mortality or all-cause mortality. There is currently no evidence to support recommending vitamins such as alpha-tocopherol, beta-carotene or retinol, alone or in combination, to prevent lung cancer. A harmful effect was found for beta-carotene with retinol at pharmacological doses in people with risk factors for lung cancer (smoking and/or occupational exposure to asbestos). More research from larger trials and with longer follow-up is needed to analyse the effectiveness of other supplements.

Respiratory organs cancer mortality in New Jersey counties and the relationship with selected demographic and environmental variables